Investment decision-makers in the oil and gas industry are faced with an extremely complex process when attempting to decide on the optimum mix of projects to pursue. Almost all firms are faced with constraints such as capital budgets, volume commitments and/or dependent and mutually exclusive projects. Likewise, attempts to include the objectives of planning, finance and engineering can make it almost impossible to find the optimum mix of projects using traditional selection methods.

Indicators from discounted cash flow analysis such as net present value, rate of return and profit investment ratios have traditionally been used to rank projects, however these rankings will not necessarily produce the best result. In addition, unless the number of potential projects under consideration is very small it can be unfeasible to evaluate all of the possible permutations. Therefore, how does one know if the best value-creating portfolio of projects has been chosen?

A recently developed technique for solving this type of problem is the use of Genetic Algorithms. Borrowing from the biological field of evolution, algorithms have been developed that can be applied to find a combination of projects that approach the true optimum, taking in to account numerous business constraints, within an acceptable time frame.

This paper describes the theory behind genetic algorithms and their application to investment decision making in the oil and gas industry. A worked example for a hypothetical company is then used to demonstrate the potential impact of using the technique.


On the surface investment management within the oil and gas industry may appear to be fairly straightforward. One may argue that the task of companies is to find hydrocarbons in economic quantities, then produce them in the most efficient manner.

Unfortunately it is not always that simple. If it is accepted that the primary goal of Exploration and Production (E&P) companies is to maximise shareholder value, managers must therefore attempt to select the best projects from those available to them. The difficulty lies in determining what are the so-called best projects when considering each one as part of the company's overall portfolio.

In an ideal world one could just choose every project available to a company that exceeded a given hurdle rate. This may be projects that are net present value (NPV) positive, achieve a certain rate of return (ROR) or a combination of those and/or other traditional economic indicators. The problem is that it is unlikely there will be no constraints on an organisation. One of the most obvious constraints is a capital budget. An organisation may not have the capacity to raise required funds internally or externally for all available projects, or they simply may not want to stretch their current resources.

There may also be conflicting aims of departments such as planning, finance and engineering. For example, a project that meets the required hurdles using discounted cash flow methods may not achieve the required short-term earnings requirements of finance. Alternatively, engineering may be measured on their ability to achieve given production levels, however doing so may require selection of projects that would otherwise be avoided.

This content is only available via PDF.
You can access this article if you purchase or spend a download.